Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images

Semantic change detection (SCD) represents a challenging task in the interpretation of remote sensing images (RSIs), with the goal of identifying change regions and extracting semantic information from bitemporal RSIs simultaneously. The recent integration of deep neural networks leveraging multitas...

Full description

Saved in:
Bibliographic Details
Main Authors: Xibing Zuo, Fei Jin, Lei Ding, Shuxiang Wang, Yuzhun Lin, Bing Liu, Yao Ding
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10737132/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850146528785596416
author Xibing Zuo
Fei Jin
Lei Ding
Shuxiang Wang
Yuzhun Lin
Bing Liu
Yao Ding
author_facet Xibing Zuo
Fei Jin
Lei Ding
Shuxiang Wang
Yuzhun Lin
Bing Liu
Yao Ding
author_sort Xibing Zuo
collection DOAJ
description Semantic change detection (SCD) represents a challenging task in the interpretation of remote sensing images (RSIs), with the goal of identifying change regions and extracting semantic information from bitemporal RSIs simultaneously. The recent integration of deep neural networks leveraging multitask learning has shown promise in enhancing SCD performance. However, there is still a challenge in improving SCD performance, specifically in designing a fine-grained network structure that can handle the two subtasks of change region localization and semantic information recognition in parallel. In this context, a novel multitask Siamese network, termed EGMS-Net, is proposed to boost the performance of SCD, which consists of three core components. First, a coarse-to-fine multitask Siamese network is constructed to obtain semantic information and change information at multiple levels. Second, an adaptive change information enhancement method based on spatial-spectral collaborative attention mechanism is proposed, which can assist the accurate localization of change regions without significantly increasing the model parameters. Third, a change information guidance module is developed to strengthen the interaction between multitask branches and reduce the difficulty of network training. Experiments on three benchmark datasets demonstrate that the proposed EGMS-Net outperforms existing state-of-the-art methods in the SCD community.
format Article
id doaj-art-236e9395ebf24a479407dd1c2dbefcb3
institution OA Journals
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-236e9395ebf24a479407dd1c2dbefcb32025-08-20T02:27:50ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118617710.1109/JSTARS.2024.348713710737132Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing ImagesXibing Zuo0https://orcid.org/0000-0002-8120-8692Fei Jin1https://orcid.org/0009-0000-4102-922XLei Ding2https://orcid.org/0000-0003-0653-8373Shuxiang Wang3https://orcid.org/0009-0004-1806-3402Yuzhun Lin4https://orcid.org/0000-0002-9560-1620Bing Liu5https://orcid.org/0000-0003-0848-8453Yao Ding6https://orcid.org/0000-0003-2040-2640Information Engineering University, Zhengzhou, ChinaInformation Engineering University, Zhengzhou, ChinaInformation Engineering University, Zhengzhou, ChinaInformation Engineering University, Zhengzhou, ChinaInformation Engineering University, Zhengzhou, ChinaInformation Engineering University, Zhengzhou, ChinaSchool of Optical Engineering, Xi'an Research Institute of High Technology, Xi'an, ChinaSemantic change detection (SCD) represents a challenging task in the interpretation of remote sensing images (RSIs), with the goal of identifying change regions and extracting semantic information from bitemporal RSIs simultaneously. The recent integration of deep neural networks leveraging multitask learning has shown promise in enhancing SCD performance. However, there is still a challenge in improving SCD performance, specifically in designing a fine-grained network structure that can handle the two subtasks of change region localization and semantic information recognition in parallel. In this context, a novel multitask Siamese network, termed EGMS-Net, is proposed to boost the performance of SCD, which consists of three core components. First, a coarse-to-fine multitask Siamese network is constructed to obtain semantic information and change information at multiple levels. Second, an adaptive change information enhancement method based on spatial-spectral collaborative attention mechanism is proposed, which can assist the accurate localization of change regions without significantly increasing the model parameters. Third, a change information guidance module is developed to strengthen the interaction between multitask branches and reduce the difficulty of network training. Experiments on three benchmark datasets demonstrate that the proposed EGMS-Net outperforms existing state-of-the-art methods in the SCD community.https://ieeexplore.ieee.org/document/10737132/Change information enhancementchange information guidancemultitask learningremote sensingsemantic change detection (SCD)
spellingShingle Xibing Zuo
Fei Jin
Lei Ding
Shuxiang Wang
Yuzhun Lin
Bing Liu
Yao Ding
Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Change information enhancement
change information guidance
multitask learning
remote sensing
semantic change detection (SCD)
title Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
title_full Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
title_fullStr Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
title_full_unstemmed Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
title_short Multitask Siamese Network Guided by Enhanced Change Information for Semantic Change Detection in Bitemporal Remote Sensing Images
title_sort multitask siamese network guided by enhanced change information for semantic change detection in bitemporal remote sensing images
topic Change information enhancement
change information guidance
multitask learning
remote sensing
semantic change detection (SCD)
url https://ieeexplore.ieee.org/document/10737132/
work_keys_str_mv AT xibingzuo multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT feijin multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT leiding multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT shuxiangwang multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT yuzhunlin multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT bingliu multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages
AT yaoding multitasksiamesenetworkguidedbyenhancedchangeinformationforsemanticchangedetectioninbitemporalremotesensingimages